|Publication number||US6980450 B2|
|Application number||US 09/882,188|
|Publication date||Dec 27, 2005|
|Filing date||Jan 24, 2002|
|Priority date||Jan 24, 2002|
|Also published as||US20040208030|
|Publication number||09882188, 882188, US 6980450 B2, US 6980450B2, US-B2-6980450, US6980450 B2, US6980450B2|
|Inventors||Suresh K. Bhate, John Cestra, Christopher B. Harrington, Leo Hoogenboom|
|Original Assignee||Inverters Unlimited, Inc.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (58), Referenced by (6), Classifications (5), Legal Events (4)|
|External Links: USPTO, USPTO Assignment, Espacenet|
1. Technical Field
The invention relates generally to switching power conversion devices, and more specifically to high power-density DC-to-AC inverters.
2. Related Art
Growing public concern about the accelerating depletion of the Earth's supply of fossil fuels (e.g., crude oil) and of other contemporary non-renewable energy sources, and energy-related environmental concerns such as pollution and global warming due to emissions, have accelerated the development of alternative energy generation technologies (e.g., fuel cells, photovoltaics, wind power, biomass, tallow trees, etc.) for generating power. Growing concerns about the reliability of local, state and national power distribution infrastructures have promoted interest in on-site power generation technologies, particularly those which are compact enough to be maintained on a private premises while providing enough power to satisfy average power requirements of a home, of a business, or of a cooperating community. Thus, a growing popular demand for technologies for independent production of electric power on the premises of individual homes and businesses, particularly solar-electric arrays and fuel cells, has increased the need for efficient and compact high-power inverters that can convert the Direct Current (DC) typically generated by devices employing such technologies, into the Alternating Current (AC) required to operate most commercially available home appliances and computer equipment. The higher the power-density (i.e., the maximum continuous power rating per unit of volume) of an inverter assembly, the less raw material may be required for its production, and the more convenient its transportation and installation, and concealment on premises can be. Some special applications for high-power inverters, i.e., for powering computers and appliances in individual homes built in remote wilderness settings, or even for powering remote extraterrestrial facilities (e.g., laboratories in orbit, or on the moon, or on other planets), may impose an even stricter necessity for true sine-wave output with a minimization of power inverter volume, or component count, and/or mass per kilowatt of rated capacity.
Pure sine-wave generating inverters generally include semiconductor transistor (e.g., Field Effect Transistor (FET), or insulated-gate bipolar transistor (IGBT)) switches controlled by sine-wave modulated pulse-width-modulated (PWM) signals. The frequency of switching is held constant (e.g., at a frequency higher than the human audio range, e.g., switching frequencies between about 20 kHz–30 kHz) while duty cycles of the alternating switches powering the primary windings are varied to produce an approximately smooth changing (alternating) potential at the secondary winding. The switches interrupt DC currents supplied to a transformer from a DC voltage source (e.g., a battery, solar array) and will generate heat during switching operation. Modified sine-wave inverters also include switches, which are used to generate pulses of alternating voltages of modulated width and fixed voltage. Both pure sine-wave inverters and modified sine-wave inverters may require a low-pass L-C output filter tuned to selectively pass the AC output frequency (e.g., 50 Hz or 60 Hz) to reduce distortion and/or remove high frequency noise. Such filters generally comprise a transformer-output filter inductor and an output filter capacitor coupled to the terminals of the secondary winding of the transformer.
Overheating of the switches often leads to damage to the switches and/or to a failure of the inverter. The heat generated in the switches is wasted energy that reduces the power conversion efficiency of the inverter and tends to limit the maximum power-output, and hence the maximum power-density and/or efficiency, of the inverter. The heat developed in the switches in an inverter generally increases at least proportionally with the frequency of the switching. The related art teaches that it is “difficult to operate a sine-wave-modulated PWM push-pull inverter at switching frequencies higher than approximately 1 kHz”. Mohan, POWER ELECTRONICS, p. 127 (John Wiley & Sons). The academic authorities also teach that PWM push-pull inverters (e.g., producing pure sine-wave output at high efficiency) are generally limited to being “Small Power” (e.g., on the order of 1 KW, e.g., 3 KW or less) inverters, compared to a full-bridge switching inverter having the same size transformer. See. e.g., Constantine Hatziadoniu, http://www.engr.siu.edu/staff/hatz/EE483/A7/sld013.htm
Therefore, although sine-wave-modulated PWM push-pull inverters require half as many switches as full-bridge (e.g., “H” bridge) inverters, they are generally unavailable as compact and high-power true sine-wave output inverters. The size of an inverter (and hence its power-density) is significantly affected by the size of its power transformer, as well as by the number and size of other inverter components such as capacitors, switch banks, heat sinks, and output filter inductors.
The invention provides configurations of various conventional and novel components to make a high power density inverter that is amenable to mass manufacturing techniques employing conventional assembly techniques such as printed circuit board fabrication, robotic component placement and wave soldering etc.
A first aspect of the invention provides a switching inverter comprising: a first primary winding connected in series to a first switch and a DC voltage source; a second primary winding connected in series to a second switch and the DC voltage source; wherein the first primary winding and the second primary winding include ribbon-like conductors each having a thickness RIBBONTHICKNESS; wherein a first coil segment of the first primary winding and a second coil segment of the second primary winding are wound coaxially around a transformer core; wherein the first winding and the second winding are approximately parallel separated by a dielectric layer, and the distance between the cross-sectional centroid of the first winding and the cross-sectional centroid of the second winding is not greater than 2×RIBBONTHICKNESS, and wherein each of the ribbon-like conductors has an aspect ratio of at least 100.
Another aspect of the invention provides an inverter comprising: a first primary winding conductor connected in series to a first sinewave-modulated pulse-width-modulated (PWM)-controlled switch and a DC voltage source, the first switch being a first composite switch including a first plurality of semiconductor switches mounted in at least one row on a first portion of printed circuit-board; a second primary winding conductor connected in series to a second sinewave-modulated PWM-controlled switch and the DC voltage source, the second switch being a second composite switch including a second plurality of semiconductor switches mounted in at least one row on a second portion of a printed circuit-board; wherein the first primary winding and the second primary winding have a minimized uncoupled inductance such that more than 100 amperes of current in the primary windings can be switched perpetually at frequencies greater than 2,000 Hz by the PWM-controlled switches.
Another aspect of the invention provides a switching inverter comprising a transformer including a first primary winding conductor connected in series to first switch and a DC voltage source and a second primary winding conductor connected in series to a second switch and the DC voltage source; wherein the first primary winding conductor and the second primary winding conductor respectively comprise a first plurality of parallel wire conductors and a second plurality of parallel wire conductors, each plurality of parallel wire conductors being intertwined with the other to form a litz-wire dual primary winding, the litz-wire dual primary winding being wound around the core of the transformer.
The present invention satisfies the need for a compact and inexpensively manufactured cooling apparatus, and methods of constructing the same, for use in the manufacture of high power-density inverters and in other devices having heat-generating chip packages. Accordingly, the inverter disclosed herein may include a compact and inexpensively manufactured cooling apparatus, such as a fluid-cooled linear heat sink in thermally conductive abutting contact with each heat-generating electronic component of a plurality of heat-generating electronic components aligned in at least one row, the heat sink having a first flat exterior surface abutted to a flat side of each component aligned in the at least one row.
The cooling apparatus may comprise a fluid conduit having substantially uniform wall thickness and a substantially uniform perimeter P formed from a thermally conductive material (e.g., which may be formed from a single tubular piece of a thermally conductive material having circumference P) and having a contact segment, wherein the contact segment has a first flat exterior surface, and wherein the first flat exterior surface has a height H1, wherein the height H1 is approximately equal to or greater than P/π. The height H1 may be equal to or greater than the height of a chip or chip package to be cooled. The invention further provides a method for making the a linear fluid-cooled heat sink having two flat sides, the method comprising: providing a linear fluid conduit having substantially uniform wall-thickness within each of two side areas and a substantially uniform crossectional perimeter P; providing a anvil having two flat surfaces inside of the linear fluid conduit and between the two side areas; pressing each of the two side areas of the fluid conduit against the flat surfaces of the anvil so as to form the two flat sides.
The present invention further provides an inverter comprising: an inductor core; a transformer including; a transformer core; a first coil wound around the transformer core; a second coil wound around the transformer core and around the inductor core. The first coil may be a “primary winding” or a “secondary winding” of the transformer. Reciprocally, the second coil may be a “secondary winding” or a “primary winding” of the transformer.
Embodiments of the present invention will be described in detail, with reference to the following figures, wherein like designations denote like elements, and wherein:
FIG. 4Ai–4A-iii depict views of an inventive high-current circuit board illustrating foil patterns for mounting multiple semiconductor switch chip packages in parallel lines to form a single PWM-controlled composite switch;
FIG. 4Ai depicts the cross-sectional view of the high-current circuit board for a composite switch along section C—C and/or section D—D of
FIG. 4Aii depicts a top view of the high-current circuit board of a composite switch wherein the linear heat sink is omitted for clarity of illustration;
FIG. 4Aiii depicts a top (x-ray) view of the bottom patterned copper foil of the high-current circuit board of the composite switch.
FIG. 4Bi depicts a cross-section of the left composite switch circuit board assembly of
FIG. 4Bii depicts a cross-section of the left composite switch circuit board assembly of
FIG. 4Biii depicts a cross-section of the left composite switch circuit board assembly of
FIGS. 7A1 and 7B1 depict side and front cross-sections of an alternative embodiment of the inverter of
FIGS. 7A2 and 7B2 depict side and front cross-sections of an alternative embodiment of the inverter of FIGS. 7A1 and 7B1 wherein the isoelectic (center-tapped) conductors have been brought into direct contact or merged;
FIGS. 7A3 and 7B3 depict side and front cross-sections of an alternative embodiment of the invention wherein the inverter is full-bridge inverter comprising a single primary winding and four PWM-controlled switches;
FIGS. 7C3 depicts a circuit diagram of the full-bridge inverter of FIGS. 7A3 and 7B3;
FIG. 7A4 depicts an alternative embodiment of the inverter of
Although the drawings are intended to illustrate embodiments of the present invention, the drawings are not necessarily drawn to scale. Various miscellaneous elements and devices that would be practical or necessary for the reliable operation of the disclosed apparatus, but which would be known understood, assumed, and supplied by persons skilled in the art as a matter of convention, have been generally omitted from the figures for the sake of clarity of illustration. Such devices omitted in the figures that may be incorporated into embodiments of the invention by persons skilled in the art of transformer or inverter manufacture, of appliance construction and/or of plumbing may include screws, nuts, bolts, brackets, braces, DC power cords, wires, wiring harnesses, insulation, dielectric (insulative) layers, sensors (e.g., current, voltage, temperature), meters (e.g., volts, amps, watts), indicator lights (e.g., “DC power on”, “AC power on” “conditions normal”), manual switches (e.g., “on”, “off”, “reset”), PWM switch-control signal generating circuits (e.g., sine-wave modified duty-cycle at greater than 10 kHz), monitoring circuits (e.g., overvoltage protection circuits, ground fault interrupt circuits, over-current trip circuits), AC electrical receptacles, hoses, tubes, pipe fittings, clamps, plumbing components, etc.
Although some embodiments of the present invention will be shown and described in detail, it should be understood that various changes and modifications may be made without departing from the scope of the appended claims. The scope of the present invention will in no way be limited to the number of constituting components, the materials thereof, the precise size and shapes thereof, the frequencies therein, the relative polarities, magnitude or ratio of voltages or currents therein, etc.
Inverters constructed in accordance with the design of
High Power-Density Inverter
High Power Density Inverters in accordance with embodiments of the present invention can be constructed in at least two different switching topologies: 1) Push-Pull (a/k/a “half-bridge”) topology as shown in
In the Push Pull topology of
The more compact the current paths outside of the transformer 216 are made, the less is the uncoupled inductance L of each primary winding. Reduction of L allows higher di/dt for a given maximum value for the Voltage across the switches. Higher di/dt in turn enables higher switching frequency. Higher switching frequency, reduces the size of the filter components on the output of the transformer to obtain smooth 60/50 Hz. sine wave output voltage (thus allowing reduction of the inverter's enclosure volume). Higher switching speed also reduces the switching losses, thus enhancing overall inverter efficiency. Higher switching speeds also reduces the size of snubber components needed to keep the voltage spikes across the switches (e.g., semiconductor switches) from reaching dangerous values.
In the “H Bridge” topology (shown in FIG. 7A3, 7B3 and 7C3), there is only one primary winding 201 within the transformer (comprised of conductor C3-C5). Each end of the single primary winding 201 is connected to two switches (end C3 is connected to switch A and switch D; end C5 is connected to switch B and switch C). One of the switch (e.g., switch A) at a given end (e.g. C3) connects that end (C3) of the winding 201 to the positive (+) terminal of the DC supply while the second switch (e.g., switch D) connected to the same end connects it to the negative (−) terminal of the DC supply. Similarly the two switches (switch C and switch B) connected to the other end (C5) of the primary winding 201 also enable the connection of that end (C5) to the positive (+) and the negative (−) terminals respectively of the DC supply. The two pairs (A paired with B) and (C paired with D) of the switches are turned on and off alternately so as to connect one end (e.g. C3) of the winding to the positive of the DC supply while the other end (e.g.,C5) is connected to the negative of the DC supply, thereby applying full DC voltage across the winding. The direction of the application of the DC voltage is alternated by alternately switching the proper switches in correct sequence, as is well understood by persons skilled in the art. During switching, only the current through the switches changes instantaneously. The current through the conductor within the transformer does not change instantaneously due to inductance of the primary winding 201. The inductance of the distinct current loops formed by the DC supply and through the switches connected to one ends of the transformer winding needs to be minimized, to minimize the L di/dt voltage spikes appearing across the switches. This applies individually to each of the two loops formed by such paths resulting from the switches connected to the two ends of the transformer winding. Minimizing these inductances results in the benefits to the design and operation of the inverter similar to embodiments of the invention employing the “Push Pull” topology.
One objective of the present invention is to configure various components in the inverter to minimize uncoupled inductances in the primary winding(s) to realize benefits and thus allow operation of the inverter at higher powers and higher power densities at higher switching frequencies and higher efficiencies than hitherto has been practicable.
The Push-Pull (Half-Bridge) Topology
The approximately uniform minimal distance between the wide surfaces (i.e., the flat sides) of the ribbon-like primary windings will typically be equal to the uniform thickness of a ribbon-like layer of dielectric (insulating) material (an insulating membrane) disposed between the surfaces of each winding. The uniform thickness of the dielectric layer should be as thin as possible without risk of mechanical or dielectric failure. Optimally, the uniform thickness of the dielectric layer will not be greater than the thickness of the ribbon conductors of the dual primary windings. Accordingly, the distance between the cross-sectional centroid of the first winding's conductor and the cross-sectional centroid of the second winding's conductor (i.e., the typical distance between the longitudinal centroidal axes of the two ribbon conductors) will typically be not greater than twice the uniform thickness of each ribbon conductor.
In embodiments of the invention, a minimally thin (e.g., less than 0.5 mm, e.g. 0.3 mm) dielectric membrane (e.g, mylar, kevlar, teflon, silicone) is disposed between minimally thin (e.g., less than 0.5 mm) conductors of the first (201-1) and second (201-2) primary windings (e.g., between conductors C6 and C5, and between conductors C3 and C4).
The physical closeness of the respective centroids of the high-aspect ratio ribbon conductors of the first (201-1) and second (201-2) primary windings throughout the inverter (200) maximizes the magnetic coupling between, and minimizes uncoupled inductance in, the current loop of the first (201-1) and second (201-2) primary windings. Uncoupled inductance in the current loops of the first and second primary windings contributes to losses and voltage spikes across the respective switches (218 and 220) during pulse-width modulated turn-off of each primary winding.
By physics, any uncoupled inductance in each primary winding is proportional to the (singly enclosed) portion of area AUNCOUPLED that is circumscribed by the current loop of that primary winding which is not also circumscribed by the current loop of the other primary winding. Thus, where the current carrying conductors (e.g., C6 and C5) of the primary windings are approximately parallel at a distance D5-6, the cross-sectional area AUNCOUPLED can be calculated as the finite distance D5-6 between the conductors of the first and second primary windings times the length over which that finite distance D5-6 apart exists. This total uncoupled area AUNCOUPLED includes area A5-6 between conductor C6 of the first primary (201-1) and conductor C5 of the second (201-2) primary plus area A3-4 between conductor C4 of the first primary (201-1) and conductor C3 of the second (201-2) primary circuit. The inductance of the loop formed by the conductor(s) C1, C2, C4 C6 through the switch 218 adds to the uncoupled inductance in primary winding 201-1. Likewise, the inductance of the loop formed by the conductor(s) C1, C2, C3 and C5 through the switch 220 adds to the uncoupled inductance in primary winding 201-2.
Area A5-6 is generally proportional to the distance D5-6 between the centroids of conductors C5 and C6 (which is approximately equal to the thickness of a dielectric membrane disposed between them plus the thickness of one such conductor). Accordingly, the thickness of the conductors C6 and C5 and the thickness of a dielectric membrane (not shown) disposed between them is to be minimized. A component of Area A5-6 is proportional to the length LSPAN of the segment of conductors C5 and C6 between the transformer 216 and switch block 230 (containing switches 218 and 220). Accordingly, the length LSPAN of that segment is to be minimized (as is depicted in FIGS. 7A1).
The total area AUNCOUPLED further includes area A3-4 between conductor C4 of the first primary (201-1) and conductor C3 of the second (201-2) primary. Area A3-4 is generally proportional to the distance D3-4 between the centroids of conductors C3 and C4 (which is approximately equal to the thickness of the dielectric membrane disposed between them plus the thickness of one such conductor). Accordingly, the thickness of the conductors C3 and C4 (which may be the same conductive ribbon as C5 and C6) and the thickness of a dielectric membrane (not shown) disposed between them is to be minimized. A component of Area A3-4 is proportional to the length LSPAN of the segment of conductors C3 and C4 between the transformer 216 and switch block 230. (The switch block 230 is the region or regions of the inverter or of circuit boards 245 therein containing switches, such as composite switch 218 (comprising switching devices 218-1 and 218-2) and composite switch 220 (comprising switching devices 220-1 and 220-2)) Accordingly, the length LSPAN of that segment should be minimized (as is depicted in
The total uncoupled inductance of each primary winding further includes the inductance due to the small current loop of conductors within the switch block(s) 230. Accordingly, current paths (i.e., loops) within each switch 218 and 220 should be effectively minimized. Embodiments of the invention provide a compact heat sink (see discussion of
The two switches (218 and 220) are controlled by two alternating sinusoidally modulated PWM signals having a high fundamental frequency (e.g. greater than 10 kHz, such as for example 28 kHz). Thus, the switching frequency of half-bridge embodiments of the inverter operates at high power and well beyond the “1 kHz” difficulty threshold noted in the academic literature quoted above. As is understood by persons skilled in the art, the Modulation Index of the PWM signal and other parameters may be set such that both switches will not be on simultaneously (i.e., to avoid short circuiting switches (218 and 220)). Dead time may be thus purposely inserted using the duty ratio of the PWM control waveform asserted at the gate(s) of each switch.
Through further minimization (i.e., shortening) of primary winding current paths (such as by reducing LSPAN as depicted in FIG. 7A1) and by minimization of underutilized space within the prototype inverter's enclosure, the dimensions of the enclosure of a 10,000 W-plus inverter having the performance and design features of inverter (200) of
Embodiments of the inverter (200) are scalable and expandable to provide continuous pure sine-wave output at any large AC power output level greater than 3,000 watts, and at AC power-densities equal to or greater than 4 Watts per cubic inch (e.g., 5.0, 6.0, 7.0 or more Watts per cubic inch). Because all the heat-releasing elements of the inverter (e.g., transformer, foil conductors, switches, capacitors, etc. can be fluid-cooled (e.g., with serpentine cooling pipes inside or affixed to their surfaces), any air-space within the enclosure of the inverter can be minimized or eliminated entirely in order to further increase the power-density of the inverter. The maximum power density of the inverter therefore is limited only by the volume of the solid components within the enclosure, such as the power transformer, the DC capacitors, fluid conduits, control circuits (each of which may be reduced to a single IC), output filter capacitors etc. Because a 60 Hz power transformer is typically the largest power-transforming component within the inverter, the power-density of the inverter can approach but not exceed the power density of the transformer itself and will be limited thereby. Variations of transformer geometry and of transformer core materials can optimize (i.e., increase) the power density of the transformer and therefore of the inverter, (with a probable increase in costs of materials etc). In a 10,000 W-rated prototype of the inverter, the transformer-to-inverter volume ratio (TIVR, i.e., the ratio of the volume of the power transformer to the volume of the entire inverter) was estimated at approximately 50%, resulting in a power density of the inverter (i.e., about 4.46 watts per cubic inch) being approximately equal to 50% of the output-power density of the power transformer (i.e., about 9.0 watts per cubic inch). In some compact embodiments of the invention such as wherein the TIVR is increased to 80% (e.g., the volume of components inside the enclosure other than the power transformer are reduced to between 10% to 15% of the volume of the transformer) the power density of the inverter may be equal to up to eighty percent of the power density of the power transformer (e.g., 7.2 Watts per cubic inch).
Logically, the performance of the inverter (200) may be easily reduced and diluted by persons skilled in the art so to obtain any lesser AC power output level (e.g., any power level less-than 3,000 W, or less than 10,000 W) and/or any lesser AC power density less than 4.46 (e.g., 4.0, 3.0, 2.0, 1.0 or less) Watts per cubic inch, while practicing the teachings of the invention.
The inverter (200) of
With Litz wire, it is possible to minimize uncoupled inductance by connecting every second alternating strand in parallel to form one winding, and connecting the remainder of the strands to form the second winding. Together, the two parallel primary windings form a litz-wire dual primary winding that is wound around the transformer's core. A wire strand of each winding may be adjacent to one or more strands of the other winding within the longitudinal segment of each winding that is wound around the transformer core. In some embodiments, each wire strand of the first primary winding may be twisted together with a wire strand of the second primary winding to form a conventional “twisted pair”. Alternatively, every wire strand of the wire strand of the first primary winding may be substantially parallel with the wire strands of the second primary winding to form the dual primary winding, the entire dual primary winding being twisted as one unit such that a first plurality of parallel wire conductors (the first primary winding) and a second plurality of parallel wire conductors (the second primary winding) are twisted together other. A Litz-wire dual primary winding may be shaped (e.g., by pressing) so that it takes the form of a rectangular or flat primary winding (e.g., the cross-sectional perimeter will have an approximately rectangular perimeter), that may be convenient to wind around the core of the transformer.
The width of a primary winding's ribbon-shaped conductor may be increased (e.g., scaled up in proportion to increased power output) without limitation (provided that the transformer shape is adjusted accordingly, or the ribbons of two windings are folded bifilarly together fit the transformer), to accommodate larger currents and larger rated-power outputs. However, the thin ribbon-thickness (RT) dimension of each foil primary conductor should be minimized (e.g., made less than 0.5 mm) and the thickness of the insulating layer (not shown) between the ribbon-like conductors minimized in order to keep the cross-sectional centroid of each primary winding as close as possible to the cross-sectional centroid of the other (adjacent) primary winding. This is done to maximize magnetic coupling between the first primary winding 201-1 and the adjacent second primary winding 201-2.
The bifilar-wound dual primary windings within the transformer 216 may be formed by stacking a first very thin dielectric (i.e., insulating) membrane upon the first primary ribbon-shaped conductor, and stacking the second primary ribbon-shaped conductor upon that first membrane, and then stacking a second very thin dielectric (i.e., insulating) membrane upon the second primary ribbon-shaped conductor, a then carefully rolling the stack up tightly around a transformer core 219, as will be understood and carried out routinely by persons skilled in the art of transformer construction.
The length LSPAN of portions of the primary winding conductors (C6, C5, C4 and C3) between the transformer 216 and the switch block(s) 230 (containing the switches 218 and 220) is minimized first by providing a printed circuit board 245 whereon each of the switches 218 and 220 (which are connected to conductors C4 and C5 respectively) is mounted in immediate physical proximity to the isoelectric (e.g., centertapped) ribbon-shaped conductors C6 and C1. As used herein, “isoelectric” means elements, wires or conductors expected to have equal electric potential (i.e., voltage), such as due to electrically conductive contact between them. The length LSPAN is further reduced by moving the printed circuit board 245 closer to the transformer, as shown in FIGS. 7A1 and 7A2. The length of the windings (and hence the magnitude of each winding's uncoupled inductance) may be further reduced by employing the arrangement of transformer-to-circuit board depicted in FIG. 7A4.
When liquid cooling of the power transformer 216 is desirable, a coil of thermally conductive tubing (e.g., copper or aluminum tubing) 203 can be wound around the transformer 216 core 219 (e.g., outside the primary coils 201) as depicted in
The cooling coil 203 may be used as an output (secondary) coil of the transformer 216. For this purpose, metal tubing of this cooling coil 203 may be coated with insulating material, so as to function as a coil of hollow insulated wire. In certain circumstances, the tube (e.g., if composed of copper or aluminum) may function as the AC-output secondary coil (e.g., 202). Alternatively, the cooling coil 203 may be used as a voltage sensor in an inverter monitoring circuit 240 (e.g., a feedback control system). Alternatively, it can be used as a power source for system auxiliaries such as a battery charger, if one is needed in an overall system of which the inverter constitutes a subsystem.
Additionally, or alternatively, any surface within the inverter (e.g., any or all of the flat sides of the transformer 216 and/or the broad surfaces of the flat primary conductors 201 external to the transformer 216) can be cooled by a fluid-cooled heat-sink, such as by applying to each such surface a serpentine fluid-carrying cooling tube 290 as is depicted in
Driver Circuit Board
Conductors C1 and C2, and C4 and C5 may be integrated with or soldered onto one or more printed circuit boards 245 (e.g., 245-R and 245-L) that carry the switches 218 and 220. Foil junctions (e.g., clamps) in conductors C4 and C5 are optionally provided to facilitate assembly and disassembly of the foil conductor stacks (e.g., for separation and removal of the transformer 216). The high-current conducting foils of conductors C1, C2, C4, and C5 facilitate very short high-current circuits paths through the switches 218 and 220 which reduces uncoupled inductances within the primary windings 201-1 and 201-2 as aforesaid.
Each switch (218 and 220) (see
FIGS. 4Ai, 4Aii, and 4A-iii depict cross-sectional, top-copper, and top view of bottom-copper views of an inventive high-current circuit board 245 (e.g., either 245-R or 245-L) having thick (e.g., TF greater than 1 mm) patterned copper foil layers for very short high-current circuits paths to and through the packaged semiconductors 260 (e.g, switches 218 and 220) which in embodiments of the inverter (200) reduce uncoupled inductances within each of the two primary windings 201-1 and 201-2 as aforesaid.
FIG. 4Ai depicts the cross-sections of j thick (e.g., TF1 greater than 1 mm) copper tongues extending conductors C4 (or C5) directly to the j drains of the nearest j packaged semiconductor switches (260-1 through 260-j). FIG. 4Aii shows a top view of these j copper tongues of conductor C4 (or C5), with the j packaged semiconductor switches drawn transparently, and the heat sink 218 (or 222) removed for purposes of illustration. FIG. 4Aii also shows the Gate Bus used to distribute the PWM switch-control signal simultaneously to the 2 j gates of all the 2 j semiconductor switches 260 (e.g., through a plurality 2 j of resistors provided to prevent auto-excitation).
FIG. 4Aiii shows a see-though top perspective view of the patterned bottom copper layer of the circuit board 245. The patterned bottom copper includes: thick (e.g., TF1 greater than 1 mm) copper foil drain links to extend a conductor C4 (or C5) from the drain of the nearest, to the drain of the farthest packaged switch 260; thick copper foil gate links, to distribute the PWM switch-controls from the gate bus (see FIG. 4Aii) to the gate of every packaged switch 260; the thick (e.g., TF2 greater than 1 mm) copper foil of conductor C1 including copper tongues which extend directly to the 2 j drains of the 2 j packaged semiconductor switches (260-1 through 260-j on each side of heat sink 218 or 220). These thick-foil patterns are only examples, and persons skilled in the art could vary the arrangement of stacked conductors C1, C2, C3, C4, C5 and C6 and the patterns thereof (See variations in FIGS. 7A1, 7B1, 7A2, 7B2, 7C1, 7C2) without departing from teachings of the disclosed invention. The within figures merely illustrate and example of foil patterns for densely mounting multiple semiconductor switch packages 260 for parallel operation to form a single composite switch (e.g., 218 or 220).
FIGS. 4Bi, 4Bii and 4Biii depict cross-sections along drain-to-drain, source-to-source and gate-to gate links, to illustrate the inventive cooperative relationship between the linear heatsink (e.g., 222) and the composite switch (218), and the high-current carrying patterned thick copper foil (e.g., tongues and links) on circuit board 245-R. This cooperative relationship facilitates the routing and switching of very high currents (many hundreds of amps, e.g. over 200 amps and over 300 amps) within a minimal footprint area on a circuit board 245 (i.e., a minimal portion of a printed circuit board), which minimizes the current loop and uncoupled inductance for each primary winding within the switch block(s) (230 as delineated in
A further aspect of the invention provides an inventive linear heat-sink and a compact parallel semiconductor mounting arrangement for forming a cooled composite semiconductor circuit (comprising a plurality of semiconductor devices connected in parallel) which is simple and economical to manufacture and assemble. The inventive cooling apparatus may be generally characterized by a linear heatsink having an interior fluid carrying conduit and at least one (e.g., two) exterior flat side for abutting with a plurality of heat-generating elements (e.g., TO-247 packaged semiconductor switches) aligned in a row and mounted on a printed circuit board. The cooling apparatus may include a metal cooling fluid-carrying conduit having two opposing flat sides, each flat side being in abutting contact with a plurality of heat-generating elements aligned in a row and mounted on a printed circuit board. Thus, the fluid conduit will be disposed between two parallel rows of heat-generating elements (e.g., as heat sink 222 and 224 are shown in the top-perspective view of
An embodiment of the inventive cooling apparatus (i.e., an apparatus for cooling a packaged or unpackaged semiconductor) comprises a length of metal conduit having one or more (e.g., two) substantially flat exterior surfaces, such as substantially parallel flat surfaces. Each flat surface of the inventive cooling apparatus is adapted to cool a plurality of abutting integrated circuit chip packages or any other heat generating components also having a flat surface. A cooled assembly including the inventive cooling apparatus can have a minimum footprint on a circuit board, and a minimum volume in an enclosure, and a minimum number of simple parts.
Embodiments of the cooling apparatus (e.g., 510) comprise a copper fluid conduit having a contact segment 511 that has two opposing flat surfaces 512 (e.g., 512-1 and 512-2) and optionally further comprising an inlet portion 520-IN and an outlet portion 520-OUT. Each flat surface 512 in contact segment 511 of the cooling apparatus 510 shall have sufficient length (L) to accommodate therealong a predetermined number (e.g., a plurality j) of discrete packaged semiconductors 565 (e.g., semiconductor chips 565, such as packaged semiconductor switches 265 within semiconductor packages 260-1 through 260-j of
Both the inlet portion 520-IN and the outlet portion 520-OUT and the contact segment 511 thereinbetween may be formed from the same fluid conduit, such as a round copper pipe having approximately uniform wall thickness T and approximately uniform diameter D and approximately uniform circumference C. The wall-thickness need not be uniform outside of the side areas of the fluid carrying conduit that shall be made into the flat sides of the heat sink. Thus, inexpensive round copper pipes with “seams” or other longitudinal imperfections formed inside or outside may be used to form the contact segment 511 of the cooling apparatus 510.
The apparatus 510 may have rotational symmetry about a longitudinal axis 505 (running from end to end through the centroid of the cross-section of the fluid carrying channel within the contact segment) and also rotational symmetry about an axis perpendicular to the plane of one of the flat sides 512. The apparatus 510 may have bilateral symmetry about a plane that is between two opposite flat sides 512 of the apparatus 510, as in the embodiment shown in
Each finite segment (e.g., 522-IN, 511, 522-OUT) of the inventive cooling apparatus 510 may have an approximately equal circumference C, as measured around the outside perimeter of cross sections along the longitudinal axis 505 of the apparatus. In particular, the circumference (i.e, the perimeter P of a cross-section perpendicular to the axis 505) CCONTACT of the flattened contact segment 511 will be approximately equal to the round circumference CINLET of the inlet cylinder 522-IN and approximately equal to the round circumference COUTLET of the outlet cylinder 522-OUT. In many embodiments, the circumferences CINLET and COUTLET of the inlet cylinder 522-IN and the outlet cylinder 522-OUT will be about equal to the round circumference C of a round pipe out of which the apparatus 510 has been formed.
In alternative embodiments the cooling apparatus 510 may be truncated at one or both ends, such that part or all of each of the inlet portion 520-IN and the outlet portion 520-OUT are omitted. For example, the cooling apparatus 510 may consist essentially of the flat sided contact segment 511, and the circumference CCONTACT of the flattened contact segment 511 will be approximately equal to the circumference C of a round pipe out of which the contact segment 511 was formed.
In various embodiments, each of two opposite flattened sides (e.g., 512-1, 512-2) of the inventive cooling apparatus 510 will have approximately equal height H (e.g, the height of first flat side H1 equals H and the height of the second flat side H2 equals H), which may be equal to or less than the total height TH of the contact segment 511 of the conduit. The height H of each flat surface 512 of the cooling apparatus 510 will preferably be at least as high as the corresponding device height dimension (HD) of a surface of a device (e.g., a chip 565 or chip package 560) that is to be cooled, thus obviating the need to provide any additional heat-distributing block or pad between the conduit 510 and the chip to be cooled. The height H may be less than, equal to, or greater than the outside diameter D of the pipe out of which the apparatus may be formed (e.g., D equals the outside diameter of the inlet cylinder 522-IN and D equals the diameter of the outlet cylinder 522-OUT). In embodiments found to be highly compact and efficient, the height H of each flattened side is equal to or greater than diameter D. Accordingly, since D=C/π in such compact embodiments, the height H will be approximately equal to or greater than C/π (i.e., C/3.14). In compact specimen of the cooling apparatus 510 that operate well to continuously (i.e., perpetually, continuing for an indefinitely long time) cool each of two banks of packaged semiconductor switches in chip packages 560 in a 10,000 W rated true sine-wave inverter powering a 10,000 watt resistive load: the height H of the (each of the two) contact segments 511 are approximately 18.75 mm; and the diameter D of the pipes and of the inlet cylinders and outlet cylinders 522 are approximately 16.0 mm (i.e, about 0.5 inches for nominally “˝ inch pipe”); and the wall thickness is approximately 0.8 mm, and the width W (i.e., distance between flat sides 512-1 and 512-2) is between about 3.5 mm and 5.5 mm (e.g., 5.0 mm), and the length of the contact segment is between 14.0 cm and approximately 17.0 cm; and the total height TH of the contact segment 511 is approximately 21.0 mm to approximately 22.5 mm; and length X of each transition portion (523-OUT and 523-IN)) can be between 12 mm and 20 mm. The length X of the transition portion can be minimized (e.g., kept less than about 130% of the diameter of the pipe) by providing a cylindrical support inside of the transition portion during formation of the contact portion.
In various embodiments, each of the heights H1 and H2 are equal to or greater than the diameter D of a pipe having circumference P, e.g., about 115% of D. A transition portion abutting the contact segment may have a length that is less than about 130% of the diameter D of the cylindrical pipe having circumference P.
In use, a flat surface 512 of the cooling apparatus 510 will be coupled to at least one heat-generating device (e.g., a chip package 560) to form a temperature-controlled assembly 500. The heat-generating device may be semiconductor chip package 560 (e.g., a TO-220, TO-247, or other flat-back chip packages) containing a semiconductor chip 565. The chip package 560 contains at least one semiconductor chip 565 or any other heat-releasing electronic or electrical element. The chip package 560 may have the form selected from the many conventional packaging structures that have a flat heat-dissipating external surface. The chip package 560 may consist essentially of a semiconductor chip 565 having wire leads (e.g., 531, 532, 533) wherein the semiconductor chip 565 is conformally coated with a layer of thermally conductive material (e.g. epoxy) formed to have a flat external surface.
The assembly 500 and its active (heat-generating) components (e.g., Power FET switch on a semiconductor chip 565) may be held near to the temperature of the cooling fluid 540 flowing within the cooling conduit 510, by mounting the chip package 560 close to a flat surface 512 of the cooling conduit 510. The shorter the distance between each heat-generating component (e.g., a Field Effect Transistor switch on the semiconductor chip 560) and a flat surface 512 of the cooling apparatus 510, the cooler (i.e., at a lower temperature near to the temperature of the fluid coolant 540) the heat-generating components can be maintained by using a fluid coolant 540 of a given temperature.
In some embodiments of the invention, a semiconductor chip 565 may be in direct physical contact with a flat surface 512 of the cooling conduit 510. In other embodiments of the invention, (as shown in
A chip package 560 may be joined to the cooling apparatus 510 in several ways. One method is bonding with the use of a heat conductive epoxy adhesive. Additionally, to maximize thermal conduction by application of pressure, or alternatively, (for purposes of facilitating easy removal and replacement of the chip package 560) a chip package can be mounted by a clamping device that applies a constant force such as commercially available binder clip springs 570 as shown installed in
The compact and linear nature of the inventive cooling apparatus 510 facilitates the cooling of chip packages while they are mounted directly on a printed circuit board 580 as depicted in
The compact nature of the cooled assembly 500 including a plurality of chip packages 560 and a circuit board 580 allows for heavier currents to be carried in shorter current paths through devices in the chip packages 560. This reduction in current path length in turn reduces resistance and resistive losses. The reduced current path also reduces circuit loop inductance, facilitating meeting very close timing requirements of high frequency circuits (such as PWM signals in high frequency inverters).
Although the inventive fluid-cooled assembly 500 is shown comprising a metal cooling conduit 560 having one or more (e.g., two) substantially flat exterior surfaces 512, the cooling apparatus 510 may have one, two, three or four or more such flat surfaces 512 in accordance with alternative embodiments of the invention. A cooling apparatus having four such flat surfaces 512 in accordance with embodiments of the invention may have a substantially rectangular cross-section with rounded or chamferred corners.
Method for Forming the Heatsink from a Round Copper Pipe
Wedge A1 and Wedge A2 may be identical in the sense that they may be mirror images of each other. Wedges A1 and Wedges A2 will have incline angles having magnitudes IA1 and IA2 respectively. Wedge B will have an incline angle equal to the sum of IA1 plus IA2. Thus, when Wedge B is mated from the opposite direction with and between Wedge A1 and Wedge A2, as shown in
Each of Wedge A1 and Wedge A2 will have two parallel flat sides (612-A1 and 612-A2 respectively) which, when combined with Wedge B inside of the pipe will provide two parallel composite flat surfaces 612 (e.g. 612-1 shown in
Wedge A1 and Wedge A2 may be (optionally) joined together as shown in
Alternative Conductor Arrangements
FIGS. 7A1 and 7B1 depict side and front cross-sections of an alternative embodiment of the inverter of
FIGS. 7A2 and 7B2 depict side and front cross-sections of an alternative embodiment of the inverter of FIGS. 7A1 and 7B1 wherein the isoelectic (center-tapped) primary conductors C6 and C3 plus C2 have been brought into direct contact or merged together (e.g., by reversing the designation and/or polarity of conductors C6 vs. C5, and C4 vs. C3).
FIGS. 7A3 and 7B3 depict side and front cross-sections of an alternative embodiment of the invention wherein the inverter is a full-bridge (i.e., “H”-bridge) PWM inverter comprising a single primary winding (C3-C5) and four (two pairs of) PWM-controlled switches (SWA, SWB, SWC, and SWD). (The well-known circuit topology of the H-Bridge inverter of FIGS. 7A3 and 7B3 is depicted in the circuit diagram of FIG. 7C3.) For a first interval of time T/2 (where T equals 1/f, where f equals the frequency of the AC output voltage, e.g., 60 Hz) a series of high-frequency PWM control pulses is applied to switch pair (A/B) which will cause them to close and open with a varying (i.e., sinusoidally-modulated) duty cycle that alternately forces and inhibits current though the primary winding (i.e., flowing from C3 through the transformer to C5). The number of pulses in the first interval T/2 is typically 100–300. During first interval T/2 the On versus Off time (i.e., duty cycle) in each cycle of the high frequency control signal is gradually shortened and lengthened to produce a half sinusoid variation of current magnitude through the primary winding and through the secondary winding. The same process is repeated during the next (second) interval T/2 by switch pair (C/D), alternately forcing and inhibiting current to flow from right to left through the primary winding (i.e., flowing from C5 through the transformer to C3) and through the secondary winding. The current through the primary winding may continue in the same direction during each half-cycle of T/2 duration, varying in magnitude gradually (i.e., sinusoidally) in many discrete steps at the fundamental (i.e., control) PWM frequency. In this manner, a complete cycle of filterably pure sine-wave alternating sinusoidal current is produced in the secondary winding of the transformer.
It will be apparent from the foregoing to persons skilled in the art that, in alternative embodiments, a four-switch H-Bridge (i.e., full-bridge) inverter may be formed by providing two circuit boards 245 (each comprising left and right halves 245-R and 245-L), each board 245 holding a pair of separate composite switches (e.g., switch A and switch C on a first board 245; switch B and switch D on a second board 245 which may be the mirror image of the first board 245), each pair of switches arranged symmetrically as in FIG. 7B1. The two boards 245 may be arranged end to end, or back-to-back and merged (e.g. switch A may be disposed upside down under switch C, and switch D may be disposed upside down under switch B) and in other configurations. The conductors C1, C2, C3, C4, C5 and C6 may be rearranged, extended and/or folded for interconnection to provide the full-bridge circuit topology.
Minimum Current Path Configuration
FIG. 7A4 depicts a side cross-sections of an alternative embodiment of the half-bridge inverter of FIGS. 7A1 and 7B1 (and alternative embodiments of a full-wave inverter as in FIG. 7A3 and 7B3) wherein the circuit board 245 has been turned 90 degrees and thereby brought closer to the transformer windings, and isoelectic (center-tapped) primary conductors C6 and C3 plus C2 have been brought into direct contact or merged together (e.g., by reversing the designation and/or polarity of conductors C6 vs. C5, and C4 vs. C3).
Integrated Transformer and Output Filter Inductor
The inventive high power-density inverter may optionally include an inventive integrated transformer core and filter core configuration wherein a filter-inductor core is integrated within the windings (e.g., coils) of the transformer, (e.g., typically within the secondary coil of a power transformer as in the case of an “output filter,” but alternatively, within the primary coil of the transformer).
The inventive ITOFI may be easily adapted to accommodate additional windings, such as a fluid-conducting cooling coil winding 903 as shown. The ITOFI of
The inventive ITOFI may be constructed and installed without separate construction and handling of a separate filter inductor. The inventive ITOFI may be manufactured as follows: first, providing a transformer core 919 and winding the primary coil (901) for placement around the transformer coil 919 in any convenient manner that such would be performed in the related art. The transformer core is designed first to obtain desired inverter characteristics. The steel core lamination material, shape, and size are determined by the requirements of the power transformer. The filter core may use the same lamination shape as used for the transformer. A filter inductor core 929 is formed (e.g., out of a stack of “E” laminations) and the secondary (output) coil is wound around the transformer core 919 and the filter inductor core 929. The predetermined amount of output filter inductance desired for achieving the desired level of output filtering can be obtained by modifying the cross-sectional area AI of the filter inductor core 929. The stack height (e.g., the number of stacked core laminates) of the filter inductor core 929 is calculated to yield the inductor core area AI required for desired level of filtering.
As an example for one particular inverter, the stack width (SWT) for the power transformer core 919 is required to be 4.5 inches, thereby requiring a stack of a certain number of each of “E” and “I” core laminations to form a three-leg core 919. And, the stack width requirement (SWI) for the filter inductor to yield the inductance required for the desired level of attenuation of the harmonics, using the same “E” and “I” laminations and the same (secondary) winding as the power transformer, is 0.5 inches.
The main difference between the power transformer's magnetics and that of the filter inductor is that, the transformer core magnetic path has zero (intended) gap in the path of the flux, while the filter inductor has a large gap to prevent saturation of the inductor core. Thus, in some embodiments, the filter inductor core 929 may be left partly or completely open at the top (e.g., to have the open shape of an “E” rather than the closed “8” shape of the transformer core 919 at some or all lamination layers). In some embodiments, the dual side return cores of the inductor core 929 depicted in
The filter inductor core 929 may be internally wider (measured perpendicular to SWI within the plane of the cross-section of
While this invention has been described in conjunction with the specific embodiments outlined above, it is evident that many alternatives, modifications and variations will be apparent to those skilled in the art. Accordingly, the embodiments of the invention as set forth above are intended to be illustrative, not limiting. Various changes may be made without departing from the spirit and scope of the invention as defined in the following claims.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US1880199||Mar 8, 1930||Oct 4, 1932||Wired Radio Inc||Inductance and coupling system|
|US2229680||May 26, 1938||Jan 28, 1941||Howard E Somes||Polyphase high frequency heating device|
|US3541487||Nov 18, 1968||Nov 17, 1970||Westinghouse Electric Corp||Electrical winding having heat exchangers between layers of the winding for cooling the windings|
|US3643131||May 1, 1970||Feb 15, 1972||Siemens Ag||Electrical device having liquid-cooled clamped disc cells|
|US3648167||Jun 1, 1970||Mar 7, 1972||Rca Corp||Fluid-cooled apparatus for testing power semiconductor devices|
|US3653433||Apr 15, 1970||Apr 4, 1972||Bbc Brown Boveri & Cie||Cooling arrangement for semiconductor valves|
|US3788393||May 1, 1972||Jan 29, 1974||Us Navy||Heat exchange system|
|US3936783||Jun 3, 1974||Feb 3, 1976||Aeg-Elotherm G.M.B.H.||Medium frequency power transformer with a secondary winding suitable for a rigid connection with an inductor through which the coolant flows|
|US4020399||Mar 12, 1975||Apr 26, 1977||Mitsubishi Denki Kabushiki Kaisha||Vapor cooling device for dissipating heat of semiconductor elements|
|US4029141||Dec 15, 1975||Jun 14, 1977||Thomson-Csf||Cooling device for components which dissipate large amounts of heat|
|US4037270||May 24, 1976||Jul 19, 1977||Control Data Corporation||Circuit packaging and cooling|
|US4093971||Dec 10, 1976||Jun 6, 1978||Burroughs Corporation||D-I-P On island|
|US4250541 *||Dec 7, 1979||Feb 10, 1981||Rca Corporation||Push-push resonant power inverter|
|US4323914||Feb 1, 1979||Apr 6, 1982||International Business Machines Corporation||Heat transfer structure for integrated circuit package|
|US4327348||Oct 3, 1979||Apr 27, 1982||Tdk Electronics Co., Ltd.||Variable leakage transformer|
|US4334267 *||Oct 20, 1980||Jun 8, 1982||Rca Corporation||Free-running push-pull inverter|
|US4381032||Apr 23, 1981||Apr 26, 1983||Cutchaw John M||Apparatus for cooling high-density integrated circuit packages|
|US4422056||Sep 28, 1981||Dec 20, 1983||General Electric Company||Integrated multi-stage electrical filter|
|US4559580||Nov 4, 1983||Dec 17, 1985||Sundstrand Corporation||Semiconductor package with internal heat exchanger|
|US4577175||Jun 27, 1984||Mar 18, 1986||Marelco Power Systems||Transformer with fluid cooled windings|
|US4675796||May 17, 1985||Jun 23, 1987||Veeco Instruments, Inc.||High switching frequency converter auxiliary magnetic winding and snubber circuit|
|US4707726||Apr 29, 1985||Nov 17, 1987||United Technologies Automotive, Inc.||Heat sink mounting arrangement for a semiconductor|
|US4721996||Oct 14, 1986||Jan 26, 1988||Unisys Corporation||Spring loaded module for cooling integrated circuit packages directly with a liquid|
|US4768581||Apr 6, 1987||Sep 6, 1988||International Business Machines Corporation||Cooling system for semiconductor modules|
|US4902942||Jun 2, 1988||Feb 20, 1990||General Electric Company||Controlled leakage transformer for fluorescent lamp ballast including integral ballasting inductor|
|US4964198||Sep 21, 1988||Oct 23, 1990||Avvid Engineering, Inc.||V-shaped clip for attaching a semiconductor device to a heat sink|
|US5001601||Jan 30, 1990||Mar 19, 1991||Grumman Aerospace Corporation||Modular cooling fixture for power transistors|
|US5023695||May 8, 1989||Jun 11, 1991||Nec Corporation||Flat cooling structure of integrated circuit|
|US5028988||Dec 27, 1989||Jul 2, 1991||Ncr Corporation||Method and apparatus for low temperature integrated circuit chip testing and operation|
|US5036384||Apr 30, 1990||Jul 30, 1991||Nec Corporation||Cooling system for IC package|
|US5210440||Jun 3, 1991||May 11, 1993||Vlsi Technology, Inc.||Semiconductor chip cooling apparatus|
|US5309319||Mar 27, 1992||May 3, 1994||International Business Machines Corporation||Integral cooling system for electric components|
|US5309979||Jun 8, 1993||May 10, 1994||Delco Electronics Corp.||Self clamping heat sink assembly|
|US5329419||Oct 21, 1992||Jul 12, 1994||Nec Corporation||Integrated circuit package having a cooling mechanism|
|US5331536 *||Nov 5, 1992||Jul 19, 1994||Opt Industries, Inc.||Low leakage high current transformer|
|US5335163||May 13, 1993||Aug 2, 1994||Scanpower||Power supply circuit with integrated magnetic components|
|US5426409||May 24, 1994||Jun 20, 1995||The United States Of America As Represented By The Secretary Of The Navy||Current controlled variable inductor|
|US5428897||Sep 29, 1994||Jul 4, 1995||Thermalloy, Inc.||Heat sink attachment assembly|
|US5430426||Sep 13, 1993||Jul 4, 1995||Tocco, Inc.||Transformer|
|US5466970||Aug 24, 1992||Nov 14, 1995||Thermalloy, Inc.||Hooked spring clip|
|US5504924||Sep 26, 1994||Apr 2, 1996||Hitachi, Ltd.||Cooling system of electronic computer using flexible members in contact with semiconductor devices on boards|
|US5549155||Apr 18, 1995||Aug 27, 1996||Thermacore, Inc.||Integrated circuit cooling apparatus|
|US5586004||Jan 20, 1994||Dec 17, 1996||Wavedriver Limited||Mounting assembly for power semiconductors|
|US5606201||May 24, 1993||Feb 25, 1997||Mannesmann Aktiengesellschaft||Fluid-cooled power transistor arrangement|
|US5725050||Sep 3, 1996||Mar 10, 1998||Thermal Corp.||Integrated circuit with taped heat pipe|
|US5731666||Mar 8, 1996||Mar 24, 1998||Magnetek Inc.||Integrated-magnetic filter having a lossy shunt|
|US5783984||Jun 16, 1995||Jul 21, 1998||Hughes Electronics||Method and means for combining a transformer and inductor on a single core structure|
|US5829516||Oct 24, 1995||Nov 3, 1998||Aavid Thermal Products, Inc.||Liquid cooled heat sink for cooling electronic components|
|US5835367 *||Jan 20, 1998||Nov 10, 1998||Industrial Technology Research Institute||Distributed plannar-type high voltage transformer|
|US5883426||Apr 18, 1997||Mar 16, 1999||Nec Corporation||Stack module|
|US5896270||May 21, 1997||Apr 20, 1999||Artesyn Technologies, Inc.||Heat sink spring clip|
|US5907479||Dec 17, 1997||May 25, 1999||Leu; Sam C.||Integrated filter forward converters|
|US5915463||Mar 23, 1996||Jun 29, 1999||Motorola, Inc.||Heat dissipation apparatus and method|
|US6014071||Jun 12, 1998||Jan 11, 2000||Matsushita Electric Industrial Co., Ltd.||Choke coil|
|US6016251||Nov 24, 1998||Jan 18, 2000||Ando Electric Co., Ltd.||Printed circuit board and cooling system therefor|
|US6028779||Nov 25, 1998||Feb 22, 2000||Hitachi, Ltd.||Power inverter device|
|US6038156||Jun 9, 1998||Mar 14, 2000||Heart Interface Corporation||Power inverter with improved heat sink configuration|
|US6046664||Mar 5, 1998||Apr 4, 2000||Century Manufacturing Company||Welding power supply transformer apparatus and method|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US7397653 *||Apr 7, 2006||Jul 8, 2008||Pv Powered, Inc.||Inverter design|
|US8243446 *||Mar 11, 2010||Aug 14, 2012||First Solar, Inc.||Photovoltaic inverter|
|US20060250760 *||Apr 7, 2006||Nov 9, 2006||Bill Taylor||Inverter design|
|US20100191489 *||Jan 28, 2009||Jul 29, 2010||Uqm Technologies, Inc.||Distributed Generation Power System|
|US20100195286 *||Feb 2, 2009||Aug 5, 2010||Rakesh Kumar Dhawan||Heat sink mechanism for internally integrated inverter hub (i3h) motor for light electric vehicles|
|US20110222242 *||Mar 11, 2010||Sep 15, 2011||Christopher Thompson||Photovoltaic Inverter|
|U.S. Classification||363/134, 363/26|
|Jun 15, 2001||AS||Assignment|
Owner name: INVERTERS UNLIMITED, INC., NEW YORK
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:BHATE, SURESH K.;CESTRA, JOHN;HARRINGTON, CHRISTOPHER B.;AND OTHERS;REEL/FRAME:011912/0070;SIGNING DATES FROM 20010516 TO 20010523
|Jul 6, 2009||REMI||Maintenance fee reminder mailed|
|Dec 27, 2009||LAPS||Lapse for failure to pay maintenance fees|
|Feb 16, 2010||FP||Expired due to failure to pay maintenance fee|
Effective date: 20091227